Tutorials, deep dives and product notes — built for developers.
Cognition’s FrontierCode 2.0 is a new 132-task mergeability benchmark. Interactive leaderboard with best-effort model scores, task pass rates, flag rates, harnesses, and actual evaluation cost per rollout. Updated October 30, 2026.
GPT-6 Luna vs GPT-5.4 Mini compared: rate card, context windows, knowledge cutoffs, and the shared Artificial Analysis harness where Luna leads the Intelligence Index 37 to 24 at 6.4x lower cost per task, while mini wins on speed. Vendor scorecard gaps stated honestly.
Across all six effort levels, GPT-6 Sol scores within a point of GPT-5.6 Sol and costs about half as much per task. Same-harness data on both, the full effort ladder, where coding improved, where knowledge work regressed, and the November price cliff.
OpenAI shipped GPT-6 Sol 90 minutes after Claude Opus 5.5, at half the token price. Artificial Analysis ran both on the same tests: Sol is cheaper to any index score up to about 44, then runs out of headroom. Every benchmark, both effort ladders and the real bill.
Opus 5.5 wins eight of the nine benchmarks Anthropic publishes against Fable 5.1 at 2.5x lower token price - but four of those margins sit inside the disclosed error bars. Includes the HAProxy same-task test, the 16-0 research-report result and the one-way thinking-block handoff.
Anthropic's Claude Opus 5.5 matches Fable 5.1 on most work at $4/$20 per million tokens: 66.4% on Terminal-Bench 4.0, 1,846 GDPval-AA Elo, and 40% lower cost per task than Opus 5. Full benchmark table with footnotes, real tester results, the four breaking API changes, and where GPT-6 Astra still wins.
Terminal-Bench 4.0: the tbench.ai public board remains led by Claude Opus 5.5 (64.8%). Adds Kimi K3.1 Pro (65.7%, Moonshot), Gemini 4 Carbon (64%, Artificial Analysis), GPT-6.1 Terra (50.3%) and MiniMax M3.1 Flash (38.1%) as separately labeled runs. Updated Oct 30, 2026.
Gemini 3.8 Flash review: DeepSWE v1.1 (73.7%), Terminal-Bench 2.1 (89.4%), full benchmark comparison vs Claude Opus 5 & GPT-5.6 Sol, real Antigravity builds with live links, and 3.8 Flash Cyber breakdown.
Comprehensive comparison of GLM 5.3 vs Claude Opus 5 with interactive radar capability chart, direct score bars, token pricing, and benchmark analysis. Updated August 19, 2026.
Interactive Terminal-Bench 3.0 leaderboard with Claude Opus 5 at 42.7%, GPT-5.6 Sol at 34.6%, GLM 5.3 at 28.3%, and all tracked frontier agent models ranked. Updated August 2026.
Claude Opus 5 vs Kimi K3: Opus 5 leads the independent BenchLM aggregate 85.88 to 79.98 and posts a 43.3–43.5% Frontier-Bench score that K3 has never even been tested on.
FrontierBench v0.1 / Terminal-Bench 3.0: adds Snorkel AI's GLM-5.3 result (32.4% ±1.5, Claude Code) alongside Z.ai's distinct 28.3% launch result. Updated October 16, 2026.